Validation of Kiremt Season Rainfall Forecast Using Gis and Remote Sensing Techniques over The Abbay Basin Of Ethiopia
No Thumbnail Available
Date
2021-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis ababa university
Abstract
Drought and flood are the most frequently occur and cause harmful impact on the socio-economic and infrastructure in Ethiopia. Theses climate related risk mainly cause due to failure of seasonal rainfall and erratic nature in distribution. This suggested that prediction of climate variability in advance of onsets of each rainfall season is the most crucial inputs for any mitigation actions. The National Meteorology Agency (NMA) of Ethiopia has been providing seasonal rainfall forecasts three times per year since 1987. This issued seasonal climate forecast needs to validate for its quality and values of the forecast. Forecast verification is an essential component in a forecasting system since it provides qualitative and quantitative measures to seasonal forecast reliability. This thesis aims to evaluate the overall performance of the kiremt season rainfall forecast based on observed stations and CHIRPS datasets over the Abbay basin issued by NMA for the last eighteen years using Geographical Information Systems and Remote sensing techniques. This verification is done by comparing observed with forecast data based on NMA monthly rainfall and CHIRPS dataset and probabilistic kiremt seasonal rainfall forecast issued from 2000 to 2018 using 38 rain-recording stations over Abbay basin.
This study uses the different attributes of seasonal rainfall forecast quality verification techniques. The verification techniques are tendency diagram, relative operating characteristics (ROC), and Ranked Probabilities Skill Score (RPSS). Based on the verification result of the kirmet season rainfall forecast, the tendency diagram reveals that forecasters tend to issue forecast probabilities for the three often favoring the normal categories as the most likely for both stations and CHIRPS datasets. This suggested that there is no difference in the quality of the kiremt season rainfall forecast when used as input stations and CHIRPS datasets. The study further showed that negative RPSS values based on station and CHIRPS datasets which indicates that kiremt season rainfall forecast are worse than the climatology predictions. RPSS values based on CHIRP datasets performed much better than station-based RPSS values.
Keywords: Kiremt, Season, Rainfall, Forecast, CHIRPS, Verification, Stations
Description
Keywords
Kiremt, Season, Rainfall, Forecast, CHIRPS, Verification, Stations